Single Node Sizing

Graylog Central (peer support)
1

Dear All,

Can anyone here tell me the correct sizing ratio of RAM a single node cluster for a small test deployment, I know there are lots of other variables that are involved. If I have 16GB available, should I do

3.0 5GB GL
5.0 GB ES
8.0 GB for OS overhead.

I had it suggested that the ES RAM could be increased so long as it was not more than 50% of system RAM, this would yield.

3.0 GB GL
7.5 GB ES
5.5 GB OS Overhead

Any one tell me which is correct and why?

Cheers

2

If you’re talking just about the RAM for the single node system and assuming that you’re only running graylog, mongodb and elasticsearch on it. I would say reserve 2GB for system, then give 8GB to ES and the rest to GL/Mongo.

don’t forget to modify the java heap.

3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.